CRATE: Clinical Records Anonymisation and Text Extraction¶
By Rudolf Cardinal <rudolf@pobox.com>.
Contents:
- 1. Overview
- 2. Package elements in brief
- 3. Reference publications
- 4. Installation
- 4.1. URLs for CRATE source code
- 4.2. Ubuntu Linux, from Debian package
- 4.3. Manual installation
- 4.4. Database and database drivers
- 4.5. Build the CRATE Java NLP interfaces
- 4.6. Configure CRATE for your system
- 4.7. Set up the web site infrastructure
- 4.8. Test the web server and message queue
- 4.9. Configure the CRATE web service to run automatically
- 4.10. Retest the web server and message queue
- 5. Databases and database drivers
- 6. Anonymisation
- 7. Natural language processing (NLP)
- 8. Configuring the CRATE web interface
- 9. Using the CRATE web interface
- 10. Ancillary tools
- 11. Troubleshooting
- 12. Technical notes
- 12.1. Resolved bugs elsewhere, previously affecting CRATE
- 12.2. Cross-platform hosting
- 12.3. Converting to SQLAlchemy
- 12.4. Installing Python 3.4 on Ubuntu 16.04
- 12.5. Transaction count always >0 for SQL Server, prohibiting CREATE FULLTEXT INDEX
- 12.6. Celery: test_email_rdbm_task() missing 1 required positional argument: ‘self’
- 12.7. SQL comments
- 12.8. webnotes.txt
- 12.9. notes_on_database_schemas.txt
- 12.10. old notes.txt
- 13. Upgrading CRATE
- 14. Common relevant TCP/IP ports
- 15. Automatic documentation of source code
- 16. Change log/history
- 17. Things to do
- 18. Abbreviations and glossary